A new model of care collaboration for community-dwelling elders ...

Volume 11, 29 April 2011 Publisher: Igitur publishing URL:http://www.ijic.org URN:NBN:NL:UI:10-1-101420, ijic2011-17 Copyright: Submitted 04 March 2010, revised 31 January 2011, accepted 21 March 2011 (Procedure frozen for 5 months – from August to December 2010 – due to personal circumstances of lead author)

Research and Theory

A new model of care collaboration for community-dwelling elders: findings and lessons learned from the NORC-Health Care linkage evaluation Corinne Kyriacou, PhD, Department of Health Professions and Kinesiology, School of Education, Health and Human Services, Hofstra University, Hempstead, NY 11549, USA Fredda Vladeck, L.M.S.W, Director, Aging In Place Initiative, United Hospital Fund, 350 Fifth Avenue, 23rd Floor, New York, NY 10118, USA Correspondence to: Corinne Kyriacou, PhD, Department of Health Professions and Kinesiology, School of Education, Health and Human Services, Hofstra University, Hempstead, NY 11549, USA, Phone: +516-463-4553, E-mail: hprcmk@ hofstra.edu

Abstract Introduction and background: Few financial incentives in the United States encourage coordination across the health and social care systems. Supportive Service Programs (SSPs), operating in Naturally Occurring Retirement Communities (NORCs), attempt to increase access to care and enhance care quality for aging residents. This article presents findings from an evaluation conducted from 2004 to 2006 looking at the feasibility, quality and outcomes of linking health and social services through innovative NORC-SSP and health organization micro-collaborations. Methods: Four NORC-SSPs participated in the study by finding a health care organization or community-based physicians to collaborate with on addressing health conditions that could benefit from a biopsychosocial approach. Each site focused on a specific population, addressed a specific condition or problem, and created different linkages to address the target problem. Using a case study approach, incorporating both qualitative and quantitative methods, this evaluation sought to answer the following two primary questions: 1) Have the participating sites created viable linkages between their organizations that did not exist prior to the study; and, 2) To what extent have the linkages resulted in improvements in clinical and other health and social outcomes? Results: Findings suggest that immediate outcomes were widely achieved across sites: knowledge of other sector providers’ capabilities and services increased; communication across providers increased; identification of target population increased; and, awareness of risks, symptoms and health seeking behaviors among clients/patients increased. Furthermore, intermediate outcomes were also widely achieved: shared care planning, continuity of care, disease management and self care among clients improved. Evidence of improvements in distal outcomes was also found. Discussion: Using simple, familiar and relatively low-tech approaches to sharing critical patient information among collaborating organizations, inter-sector linkages were successfully established at all four sites. Seven critical success factors emerged that increase the likelihood that linkages will be implemented, effective and sustained: 1) careful goal selection; 2) meaningful collaboration; 3) appropriate role for patients/clients; 4) realistic interventions; 5) realistic expectations for implementation environment; 6) continuous focus on outcomes; and, 7) stable leadership. Focused, micro-level collaborations have the potential to improve care, increasing the chance that organizations will undertake such endeavors.

Keywords collaboration, linkage, aging, chronic illness, community-based

This article is published in a peer reviewed section of the International Journal of Integrated Care

1

International Journal of Integrated Care – Volume 11, 29 April – URN:NBN:NL:UI:10-1-101420/ijic2011-17 – http://www.ijic.org/

Introduction and background Providing care for elderly individuals with chronic illnesses in the community requires a model of service delivery that takes into account both physical health and social health needs. However, packaging care in this way does not fit into existing service or reimbursement structures in the United States, and there are few financial incentives that encourage coordination. In fact, funding and reimbursement systems tend to discourage cross setting integration [1, 2]. While there are a small number of innovative programs that have pooled resources and coordinated care for discrete populations across service systems (e.g. Social HMO, PACE, MMIP, PROCARE), they tend to be highly complex, limited in scope, costly, and large scale evaluations have found mixed results. Indeed, several of these programs are no longer in operation—or exist in an extremely limited fashion—due to their inability to show positive outcomes and/or due to their poor cost-benefit ratio [3]. Furthermore, most social service and health care managers are not trained in developing effective multidisciplinary programs with other types of providers. The organizations within which they work tend to be departmentalized and fragmented along functional lines. Lack of coordinated care can negatively affect access to high quality, appropriate care, putting at risk seniors’ physical and mental health, quality of life, and ability to stay in the community.

The rise of NORC-SSPs About 20 years ago, social service providers in New York State developed a model of care aimed at overcoming service fragmentation and its potential risks for community dwelling elders [4]. The first step towards the development of a new model of care was the realization that there were age-integrated housing developments and neighborhoods throughout New York City where large numbers of elderly persons were residing and in need of supports and services where they were living. Naturally Occurring Retirement Communities (NORC) became the natural home for Supportive Service Programs (SSP). The first NORC-SSP was established in 1986 at Penn South Houses in New York City. In 1995, New York State endorsed the model by providing funding to create 14 NORC-SSPs; New York City followed suit in 1999. Today 54 NORCSSPs operate in New York State with 43 of them in New York City. Together these programs serve multiage communities in which more than 67,000 seniors live [5, 6]. Over the past 10 years, the NORC program concept has spread to more than 40 communities in 25 other states through the use of federal earmark dollars, and several states are piloting state-wide ini-

tiatives. NORCs have evolved from primarily vertical arrangements (i.e. in high-rise, city-based apartment buildings) to both vertical and horizontal arrangements (the latter referring to suburban-based, single family homes). In October 2009, Community Innovations for Aging-in-Place national demonstration began under the auspices of the United States Department of Health and Human Services – Administration on Aging to test models and approaches, including a strong emphasis on the NORC program concept. NORC-SSPs unite housing entities, health and social service providers, residents and other community stakeholders, government, and philanthropic organizations to provide a wide range of services, early interventions, and meaningful activities for seniors in communities where they live. The NORC-SSP model represents a significant departure from the current service delivery system based on functional deficits. From program development to the definition of client, the model expands the role of older people in their community from recipients of services to active participants in shaping their community as ‘good places to grow old’ [4]. The model also assumes quite different approaches to defining and therefore financing services, and to collaborations among health and social service providers. New York NORC-SSPs are distinguished by the following hallmarks: 1) They are based on communityidentified challenges to aging-in-place; 2) Residents themselves play a vital role in the development and operations of NORC-SSPs; 3) They are financed through public-private partnerships that combine revenues and in-kind supports; 4) Their programs promote independence and healthy aging by engaging seniors before a crisis and responding to their changing needs over time; and, 5) Eligibility for services and programs is based on age and residence in the NORC, rather than on functional deficits or economic status, and the mix of services available is resident-specific, not program specific.

Integrating health and social health care in NORC-SSPs All NORC-SSPs provide social work services; indeed, in most NORC-SSPs in New York City, the lead agency is a social services agency. Most NORC-SSPs in the city have a health care partner as well; the partner may be a certified home health agency, nursing home, or hospital. Educational and recreational activities and volunteer opportunities are diverse and designed to engage as many community residents as possible. Although organized and managed by the professional staff, many classes or activities are led by the seniors


2


themselves. Because success depends on the extent to which a NORC-SSP reflects the strengths, interests, and aspirations of community residents, thorough assessment, extensive and ongoing outreach, and the ability to understand and adapt to changes in the community over time are essential. New York’s NORC-SSPs have developed various governance structures in order to manage the complex partnerships of housing corporations, social service agencies, health care providers, government agencies, and the residents themselves. NORC-SSPs work hard to make the collaboration among these diverse partners viable; strong leadership is key, as is an ability to redefine institutional boundaries and relationships. Despite ongoing growing pains, the New York NORCSSP experience has demonstrated that it can be done: public programs, service delivery organizations, and communities themselves can come together to create and operate totally new forms of senior services, organized around the seniors and their communities, which can make a positive and palpable difference in individual lives [4]. There is a growing body of literature that recognizes the importance—and potential—of health and social service organizations working together to improve health outcomes in communities [7–13]. The majority of well-known community health improvement programs have targeted a single disease and have been organized as ‘top-down’ initiatives, often with a university-

Self care NORC PROGRAM

• Empowerment • Self-advocacy • Lifestyle choices

based research group leading the effort [14]. While the logic of implementing health promotion interventions within the community is clear, evidence demonstrating impact on targeted health outcomes is sparse. Health improvement through community-based interventions remains a challenge and calls for new theories, models and methods [14]. The NORC-SSP model redefines the classic chronic care approach by recognizing that medical care, community care and self care are equally important components of comprehensive chronic care management. Departing from the classic institution-centric model, this Community Chronic Care Model (see Figure 1) underscores the critical role of the NORC-SSP in integrating the essential components of successful agingin-place for chronically ill individuals residing in the community.

The NORC-SSP Linkage initiative In 2002, two New York City-based funders, the United Hospital Fund1 and The New York Community Trust2 initiated a demonstration project, termed the NORC-SSP Linkage Project, which included coordinated grantmaking to five project sites to design and implement collaborative approaches3 to information sharing, tracking, and producing outcomes for community-dwelling elders. The grants also allowed for the provision of technical assistance to grantees by

Community care

Medical care

• Environment • Resources • Support

• Diagnosis • Treatment • Disease management

United Hospital Fund 2008

Figure 1. Community chronic care model.

1 The United Hospital Fund (UHF) is a non-profit health services research and philanthropic organization whose mission is to shape positive change in health care for the people of New York. In 1999, UHF established the Aging-in-Place Initiative to foster the development of new models of care supporting the health and well-being of older people living in the community. While New York is the focus of this work, its impact and relevance is national. For more information, see www. uhfnyc.org. 2 The New York Community Trust (Trust) is one of the oldest and largest community foundations in the United States. Founded in 1924, the Trust operates more than 2000 charitable funds and focuses its efforts in the following four areas: Community Development and the Environment; Health and People with Special Needs; Education, Arts and Human Justice; and, Children, Youth and Families. For more information, see www.nycommunitytrust.org. 3 Also called ‘comprehensive wrap around’ approaches in the literature [26].


3


Fund staff, and a robust evaluation of the individual sites and the project as a whole. The Linkage Project was developed with the recognition that integrating health and social services is both complex and necessary to promote successful aging-in-place. By design, NORC-SSPs attempt to increase access to care and enhance care quality for aging residents through a complex web of services that include community outreach, needs assessment, service coordination, service provision and ongoing client monitoring. Given that the overarching goal of the NORC-SSP is to enable residents to ‘successfully age in place,’ a focus on both social services and population-based health care is essential. Since their inception in 1986, NORC-SSPs have partnered with local health care providers to address both individual and communitywide health issues. However, while NORC-SSPs create greater collaboration between social service and health providers, each of these sectors continue to operate along functional lines, and on a reactive basis. Indeed, the realization that health and social service providers were essentially speaking ‘at’ each other rather than ‘with’ each other about managing chronic illness led the funders to undertake the Linkage initiative. The integration of health and social services is therefore defined more broadly in the Linkage project than in traditional NORC-SSP-health care partnerships. Health and social services are not seen as supplemental to each other nor connected merely by referral channels, but rather defined by multidisciplinary needs assessment, targeted program planning, care planning and management, and ongoing follow-up. Although the concept of service integration is more comprehensive in the Linkage project, the vehicle is less intense than a ‘partnership’ in the formal sense. The goal of The Linkage Project was to foster the development and testing of new models of collaborative and coordinated problem-solving, models that involve ‘micro-collaborations’ using simple, familiar and relatively low-tech approaches to sharing critical patient information among collaborating organizations. Participating NORC-SSPs and health care providers worked together to identify health conditions in the community that could benefit from more systematic communication and information exchange as well as targeted programming. Targeted conditions were: falls; discharge planning and medication management; depression; and diabetes. Gaps in services were identified and discrete mechanisms or strategies—that is, linkages—were designed to fill those gaps. A central goal was for the linkages to become part of routine practice at each collaborating organization. Rather than dedicating a staff member to deal with individual health problems as they arise, the Linkage Project

defines health as multi-faceted and focuses on population-based approaches to prevention, health promotion and multidisciplinary care planning.

Methods Conceptual framework for the NORC-Health Care Linkage evaluation Evaluations of initiatives implemented across multiple settings, each with different partnership relationships, different population targets and different interventions are highly complex. One major problem in evaluating multi-site community-based projects is that the overall goals and objectives are difficult to conceptualize. While individual projects may claim to share a common vision, they function in very different environments and often approach the same problem in different ways [15]. There are many theories that help explain collaborative action and the factors likely to influence it. They come from the disciplines of organizational behavior, sociology, political science and economics. However, most classical theories do not complement each other well, and community-based health care initiatives require theoretical guidance that takes into account the complexities of creating partnerships in addition to the challenges of explaining health behavior change within a highly complex and variable external environment. Theories of action (TOAs) explain how a program is expected to get from conditions at baseline to a desired future; thereby bridging strategic planning and evaluation [15]. Clarifying underlying assumptions can help to articulate and operationalize hypotheses, research questions, variables of interest and appropriate data collection instrument. TOAs can also help evaluators better understand when expected short-term and longterm outcomes might be observable by examining the order and various levels of anticipated effects [15]. The most effective TOAs are co-generated by evaluators and partnership representatives. Working collaboratively also fosters ownership to the components of the theory and enables individual organizations to develop their own theories. Since the purpose of evaluation of complex community initiatives is to facilitate their improvement and effectiveness [16], programs must be guided by more specific ‘treatment theories’ that will explain how interventions will reach the target population in sufficient ‘dosage’ to be detectable [14]. Treatment theories attempt to explain how inputs translate into outputs; how programs plan on producing anticipated effects. Moving from the abstract identification of an overall vision and ultimate program goals to the nuts and


4


bolts of designing and implementing interventions that will enable those goals to be achieved requires linking theory to practice. While TOAs are more concrete than classical theories of organizational behavior or social change, they still need to have their components defined and linked to realistic indicators, measures and timeframes. The logic model provides a tool for conceptualizing the relationships between short-term outcomes produced by programs, intermediate system impacts and long-term community goals [17, 18]. The logic model is described as a logical series of statements linking a condition(s) in the community, the activities that will be employed to address the condition(s), short-term outcomes resulting from activities and longterm impacts likely to occur as multiple outcomes are achieved [19]. The logic model treats each partnership as a separate case study recognizing that the communitydriven approach results in a wide range of goals and objectives and program outcomes [18]. The value of the logic model is its ability to consider connections between conditions, activities, outcomes and impacts [19]. Conditions should reflect concerns of local populations that can be realistically addressed through the interventions defined in the activities component. Similarly, activities should lead in a logical sequence to short-term outcomes and such outcomes should contribute to the achievement of longer-term system impacts and/or community goals. Looking at outcomes in this hierarchical way provides a framework for the Linkage evaluation. Each of the participating Linkage sites developed a TOA and a Logic Model. Empirical basis for testing effectiveness relied primarily on case methodology, although other qualitative approaches were utilized to overcome the bias inherent in any one method, and to increase validity because different methods highlight different aspects of the experience [16]. The logic model enables us to look at individual sites and collective experiences and compare outcomes across common dimensions [20]. The Linkage evaluation followed the lead of other multi-site, community-based evaluations [13–15, 20] in recognizing the importance of changing organizational behavior as a prerequisite for changing client/patient behavior. It was hypothesized that effective service linkages could only occur after the collaborating organizations saw each other as offering essential services that could help them better care for their clients/patients. To realize this value, organizations first needed to know that each other exist, understand what services each provide, how these services fill gaps in care that negatively affect their clients/patients and how the integration of these services could be realistically achieved.

Therefore, immediate outcomes, the first level, were specified as those that demonstrate increased awareness about the other sector providers and increased knowledge of the target population; intermediate outcomes, the second level, were specified as those that demonstrate changes in practice and behavior; and, long-term/distal outcomes, the highest level, were specified as those that demonstrate changes in health status. Immediate and intermediate outcomes may cumulatively, but not necessarily directly, lead to changes in more distal outcomes [10]. The evaluation approach offered Linkage sites great flexibility in designing, implementing and evaluating their own interventions, with guidance from the funders and external evaluator. The importance of evidencebased program planning was emphasized from the outset. Sites were encouraged to see the value in tracking their own progress for both accountability purposes and future programming. This innovative initiative aimed to build capacity so the participating sites could continue to apply these new skills long after the funder and external evaluator had gone back to their respective professional homes. This article presents findings from the evaluation of the demonstration sites’ experience over two years (2004– 2006); year 1 involved planning and year 2 implementation of the Linkage interventions. The evaluation looked at the feasibility and quality of linkages, as well as the impact of the linkages on selected care process and clinical outcomes. Using a case study approach, incorporating both qualitative and quantitative methods, this evaluation sought to answer the following two primary questions: 1) Have the sites created viable linkages between the participating organizations that did not exist prior to the Linkage initiative; and, 2) To what extent have the linkages resulted in improvements the key variables of interest: knowledge and awareness of partner services and target population; communication among partners; shared care planning; continuity of care; and outcomes of care.

Data collection strategies and instruments Immediate, intermediate and distal outcomes were tracked using tracking forms, client surveys, stakeholder interviews, chart reviews and periodic site visits. Intensive technical assistance enabled the participating sites to take an active role in data collection and analysis. Comprehensive data guides were produced for each site with specific steps for implementing interventions, tracking progress and collecting data. Using


5


a consensus process, the data guides outlined data sources, responsible parties and timelines. Participating sites submitted quarterly reports including both a narrative discussion and data on progress. Outcomes tables were constructed for each site per quarter based on these reports and supplemented with additional data generated from other sources (i.e. interviews, site visits). Tables tracking overall progress regarding implementation of activities and outcomes were also compiled. Three months into the implementation year, and then again at the end of the implementation year, staff involved in the planning and implementation of the Linkage projects at each site—defined as Stakeholders—completed a web-based survey assessing the collaboration. The evaluator conducted site visits and follow-up telephone interviews with program staff at several points throughout the grant period to collect additional information on the process and to provide technical assistance with data tracking, as needed. Process measures, including level of community participation, level of site participation, planning products developed, use of financial and human resources, and services provided were examined.

Description of Linkage participants, theories of action and interventions The Linkage project supported the efforts of four NORCSSPs4 operating in New York City, in establishing service linkages with health care providers to improve care. All four sites brought together the NORC-SSP with that community’s key health care provider—hospital, primary care clinic, or voluntary community physicians to collaborate on addressing health conditions that could benefit from a biopsychosocial approach—conditions that require effective co-management of physical, psychosocial and environmental factors. Table 1 describes the four Linkage communities by geographic location, type of housing, number of residents, staffing structure, existing health care partner, and linkage focal point. Each Linkage site focused on a specific population, addressed a specific condition or problem, and created different linkages to address the target problem. Focusing on the common goals of increasing access to care, improving continuity of care, and improving care quality and outcomes for community-dwelling elders by integrating health and social services, each Linkage site worked with the funders and evaluator to develop a local theory of action, strategies for creating linkages among select providers, and interventions to test the theories. 4 Initially five sites were funded; however, after the first year, one of the sites dropped out of the initiative.

Mid-Manhattan site The Mid-Manhattan site sought to improve emergency room diagnosis, treatment and discharge planning by strengthening relationships between the NORC-SSP, the Emergency Department (ED) at the participating hospital, and local pharmacies. The initial theory behind this linkage was that with immediately available, comprehensive information about medications, ED physicians would be better able to care for seniors in crisis, hospitals would be able to improve the discharge planning process, and the NORC-SSP would be better equipped to support medication compliance after discharge, perhaps preventing inappropriate rehospitalizations. The Mid-Manhattan NORC-SSP already partnered with the local hospital for the on-site nurse; however, their linkage intervention expanded this limited relationship by initiating an intervention where the hospital emailed5 the NORC-SSP a daily list of discharges from within the NORC catchment area. For the first time the NORC-SSP nurse and social worker would be able to proactively follow-up with all NORC residents that had been seen in the ED. NORC staff could now address the problems that triggered the ED visit, assist with the transition back to the community for those that had been discharged following an inpatient stay, and track health trends within their population for the purposes of planning health related programming and services. The hospital would expand its reach as well. Delayed, absent or poor follow-up post-discharge can lead to inappropriate readmissions to the hospital. Building upon the ED discharge collaboration, the Mid-Manhattan NORC-SSP reached out to local pharmacies to create and manage an electronic medical record, called MyMeds, that would be shared by the NORC-SSP, the ED and local pharmacies. This intervention was based on a theory of action that said the more providers who know about all the medications a resident is on, the less chance there is for errors and treatment delays and the more likely it will be that the resident will adhere to the treatment regimen and experience positive outcomes. The NORC-SSP held informal training and education sessions with pharmacies as they came on board about the MyMeds process and purpose. ED physicians were educated about the program during staff meetings. Initially, residents were recruited to enroll in the MyMeds program upon discharge from the ED, although over time additional residents who heard about the program by word of mouth were also enrolled. Medication profiles were created for all enrollees and uploaded at 5 The NORC nurse in this case was an employee of the hospital and therefore entitled to the information without violating HIPAA rules.


6


Table 1. Description of funded linkage communities NORC-SSP site

NORC community

Staffing

NORC health care partner (existing)

Linkage focal point (new)

MidManhattan

A coalition of a public housing complex and a moderate-income cooperative on Manhattan’s Upper West Side where more than 800 of the approximately 3200 residents are seniors. Senior residents are primarily Black and Hispanic (76% combined). NORC-SSP established in 2000

Two and a half full-time social workers, one full-time nurse

Local hospital

Emergency Department of local hospital partner, and local pharmacies

Lower Manhattan

A public housing project on Manhattan’s Lower East Side with 27 buildings, 3000 residents, 860 of whom are seniors. The senior population is diverse: 59% Hispanic; 22% Asian; 14% White; 5% Black. NORC-SSP established in 1993

Four full-time social workers, one part-time nurse (3 day/week)

Certified home care agency

Local primary care clinic

Queens

A moderate income garden apartment cooperative in northeast Queens, home to more than 4000 residents, 1000 of whom are seniors. Senior residents are overwhelmingly White. NORC-SSP established in 2000

Two full-time social workers, and a 75% time nurse (4 days/week)

Local hospital

Community physicians affiliated with local hospital partner

Brooklyn

A large, primarily low-income rental housing complex in an isolated part of southern Brooklyn, with 46 buildings, 14,000 residents, 2700 of whom are seniors. The senior population is diverse: 44% Black; 41% White; 15% Hispanic and a growing Russian population. NORC-SSP established in 2000

Three full-time social workers, one full-time nurse

Local hospital

Community physicians

their pharmacy of choice. Enrollees were educated about the purpose of the program and the importance of medication compliance. Bracelets specifying current medications were offered to all enrollees. Lower Manhattan site The Lower Manhattan site theorized that if the NORCSSP worked together with the local primary care center on managing care for Hispanic residents with diabetes, care outcomes would improve. The linkage sought to connect these two organizations through the sharing of information and coordination of efforts. Initially, paper forms with diagnostic, treatment and appointment information were faxed back and forth; over time a more technologically sophisticated approach—a shared, electronic patient database—was established. The NORC-SSP already partnered with a certified home care agency for the on-site nurse and this new transfer of information would enable her—and the social work staff—to systematically determine who among their residents were being treated for diabetes and how best to help them manage their illness. While the NORC-SSP staff had long been trying to monitor clients with diabetes, they were not always aware of prescribed treatment regimens, dietary restrictions, scheduled follow-up appointments, or how to identify signs of poorly managed symptoms. The shared database would enable NORC-SSP staff to use the health status data entered

by the primary care center staff to determine where to focus their efforts (e.g. symptom education, assistance with shopping or transportation, home care referrals for nutrition education, diet planning, glucometer training, medication administration or monitoring). The primary care center, on the other hand, specialized in diabetes care—indeed it had an ongoing Diabetes Collaborative—but was disconnected from patients as soon as patients left the medical office and therefore limited in how it could intervene in the patient’s dayto-day disease experience and follow-up care. The primary care center-based physicians could use the social status data entered by the NORC-SSP staff to better understand their patients’ living situation, social health needs and economic status, thus enabling a more tailored and comprehensive treatment plan and minimizing the risk of decline associated with poorly managed symptoms among patients with diabetes.6 The NORC-SSP held training and education programs for the primary care center staff in an effort to increase 6 The Linkage Initiative, with its major objective of cross sector sharing of patient/client information, began just as the Health Insurance Portability and Accountability Act (HIPAA)—with new rules for how and when health providers could share patient information—became a reality for all health care providers [24]. The funders provided the participating sites with the consultation services of an elder law attorney who had been trained in HIPAA rules and regulations. Armed with clarification about how to and the conditions under which projects could legally share protected information, all sites successfully negotiated the new HIPAA rules.


7


knowledge of partner capabilities and services. To empower seniors to become more active participants in the management of their illness, one-on-one counseling sessions (at the NORC-SSP or in the home) as well as group visits (at the primary care center) were implemented. Queens site The Queens site theorized that if the NORC-SSP worked closer with community physicians they could decrease falls among their residents.7 They recognized that each provider had important pieces to the falls prevention puzzle—the NORC-SSP had information on the home and residents’ day-to-day lives, the community physician had information on the residents’ medical conditions and medications. Putting the puzzle pieces together in a proactive, collaborative way was expected to optimize resident care and decrease preventable falls. The Queens site already partnered with the local hospital for the on-site nurse which helped to facilitate the linkage with hospital-affiliated community physicians. The system of communication used by the Queen site involved the sharing of community-based risk assessment results with community physicians. Home visits to assess the environment for falls risk and to screen for other falls risks (e.g. trigger drugs, dementia) were conducted on all residents. A five-point risk screening tool (age, history of previous falls, use of assistive devices, use of 5+ medications, and ability to ‘get up and go’) identified those in need of a multifactorial assessment and comprehensive, integrated, intervention approach. At risk residents were assessed by NORC-SSP staff using the Hartford Scale Falls Risk Assessment [21], and the score faxed to the resident’s physician in the form of a short and concise consultation letter. The style of the letter was similar to the familiar home health referral letter making it easy for physicians to request specific services from the Queens-based NORC-SSP (including additional information) or make a referral for a physical therapy evaluation, durable medical equipment, home care, or other services. Physicians were able to respond to the letter by checking off desired actions, signing his or her name for orders and faxing back the letter. The NORC-SSP would then implement the actions, where appropriate, or facilitate referrals. The goal was to shift communication between the NORC-SSP and community physicians from reactive and crisis-driven to proactive, thus expanding options to optimize patient care. 7 The Queens site chose to focus on falls after seeing data from the local hospital that showed that 80% of the seniors seen in the emergency room had experienced a fall.

Comprehensive presentations were made to community-based physicians to educate them about the NORCSSP, the Falls Risk Reduction initiative and falls among the elderly in general. Resident-centered falls training and education programs—group and one-on-one—for all seniors at the Queens NORC-SSP were implemented. These programs were aimed at increasing knowledge of risk factors for falls, improving self care and patient confidence, and reducing risk factors for falls. Brooklyn site The Brooklyn site theorized that depression among the elderly goes untreated because elderly residents rarely speak to their primary care physicians about their emotional concerns, and because primary care physicians are not well connected to specialty geropsychiatric services, nor sufficiently aware of community-based mental health services. Therefore, this site sought to establish a proactive, shared care planning approach where mental health assessment data was collected by the NORC-SSP nurse and shared with community physicians for use in identification of problems, diagnosis and treatment. Mental health assessment data was gathered using a standardized depression assessment tool, the Hamilton Depression Rating Scale [22, 23], and a resident passport—a paper record carried to all visits by the resident—was established as the vehicle to share information among providers. To ensure the community physicians had the resources they needed to use the assessment data comprehensively, the NORC-SSP offered information on community-based mental health services and created opportunities (i.e. professional programs) to bring geropsychiatrists from the local hospital and community physicians together for consultative purposes. The Brooklyn site already partnered with the local hospital for the on-site nurse and this enabled them to facilitate the linkage with the geropsychiatric unit. It was expected that providing mental health assessment data would assist physicians in both diagnosis and treatment, and involving the NORC-SSP nurse would improve the facilitation of referrals to outpatient mental health providers (where possible and when necessary). The NORC nurse would also serve as the liaison between the specialist and primary care provider, further expanding the linkage between social services and health care. Education programs for providers and residents were designed to increase knowledge of depression signs, symptoms and treatment strategies, as well as increase awareness of treatment options, services and strategies for improving communication.


8


Table 2. Primary linkage interventions: systematic sharing of information NORCSSP site

Target condition

Description of systematic sharing of information

MidManhattan

Transition from ED to community

• Daily email list of ED discharges within the NORC-SSP catchment area • Electronic medication record housed by and updates by local pharmacies

Lower Manhattan

Diabetes

• Shared Electronic Patient Database with diagnostic, treatment and appointment information

Queens

Falls

• Consultation letter from NORC-SSP to PCPs with results from the Hartford Falls Risk Assessment Protocol and recommendations for treatment and/or referral

Brooklyn

Depression

• Shared Client Passport with results from the Hamilton Depression Screening Protocol and notes on all medical and social work visits

Although initially designed as formal educational programs, the NORC-SSP modified its approach once it was clear that lectures and group sessions would not work for providers nor residents. Information and training was thus provided in more informal, face-toface meetings.

Common interventions The primary intervention across all sites was the systematic sharing of specific information among partners that had in the past only shared such information on an as-needed, ad hoc basis, or crisis-driven basis. In most cases this sharing of information was electronic, which had the added benefits of speed, accuracy and documentation. Table 2 reviews how each site formulated their primary interventions. Secondary interventions at all sites included professional education and training on the linkage intervention for professional staff at partnering organizations, and education and outreach associated with the target conditions for participating NORC residents. The following section reports on the evidence gathered that demonstrates the impact of these interventions.

Findings Evidence of immediate, intermediate and distal outcomes Evidence collected suggests that immediate outcomes were widely achieved across sites: knowledge of other sector providers’ capabilities and services increased; communication across providers increased; and identification of target population increased. Intermediate outcomes were also widely achieved: shared care planning increased and continuity of care was enhanced. Furthermore, preliminary evidence sug-

gests that even the more distal outcomes of improvements in disease outcomes were selectively achieved as well. Each site decided upon a recruitment and enrollment strategy that worked best for them. Each aimed to enroll between 50 and 100 at-risk residents using a convenience sampling strategy. The definition and determination of risk differed by site and program focus. At the Mid-Manhattan site, residents recruited upon discharge from the ED, were enrolled (n=100). At the Lower Manhattan site, Hispanic residents who were also patients of the partnering primary care center were identified, recruited and enrolled (n=39). At the Queens site, a risk screen conducted by NORC-SSP staff during routine home visits identified residents at high-risk for falls (n=100). In Brooklyn, residents known to the NORCSSP staff to have emotional problems were screened for depression risk, and those found to be at risk were enrolled (n=45). All NORC-SSPs provided some sort of professional education to increase awareness of their services and capabilities among their new Linkage health care partners. Surveys and interviews with staff—and in some cases residents—assessed whether these educational initiatives were effective (Table 3). The two sites where the NORC-SSP was collaborating with a health care organization were able to demonstrate increased awareness. Indeed, at Lower Manhattan, 100% of resident enrollees queried at the conclusion of the implementation year reported that the primary care center already knew they were NORC program clients at the time of their last visit and/or asked them if they were a client of the LowerManhattan NORC-SSP (data not shown). The two sites where the NORC-SSP partnered with community-based physicians did not conduct surveys to determine whether there was increased awareness (there was a concern that such surveys would be too burdensome and the NORC-SSPs did not want to risk losing the community physicians’ interest in participat-


9


Table 3. Increased awareness of partner services Site

Evidence

MidManhattan

• Seventy-one percent of ED staff correctly identified the NORC program, its services and client eligibility • One hundred percent of ED staff correctly reported the MyMeds program purpose • Eighty-six percent of ED staff correctly identified MyMeds partners • Ninety-three percent ED docs were able to correctly report how to identify a MyMeds member

Lower Manhattan

Seventy-four percent of primary care physicians working in diabetes at the primary care center were aware of the Lower Manhattan NORC, and 63% of the joint diabetes Linkage project

Queens

• Twelve formal presentations were made by the NORC-SSP for community physicians (3–4 per quarter) • On average, 24 community physicians attended each presentation

Brooklyn

Anecdotal evidence (i.e. increases in documented communication between NORC-SSP and physicians during the Linkage project) suggests increased awareness

ing in the Linkage initiative). However, anecdotal evidence suggests that despite the challenge of engaging community physicians, awareness of NORC-SSP activities increased. Both sites used multiple strategies (i.e. formal presentations, office visits, newsletters, direct mailings) to inform community physicians about the NORC-SSP in general, its client population, pro gress on the Linkage, and the health condition of interest. The Queens site was more successful than the Brooklyn site in getting community physicians to attend formal presentations. This is likely due to the fact that all community physicians targeted by the Queens NORCSSP were affiliated with the hospital where the presentations were held (this hospital also had an established relationship with the Queens NORC-SSP). The Brooklyn NORC-SSP also had an established relationship with the partnering hospital; however most community physicians targeted for the Linkage project were not affiliated with that hospital. Nonetheless, more informal contact between the NORC-SSP and the physicians— in the form of office visits by the NORC-nurse—led to increased awareness of partner capabilities and services, as per NORC-SSP staff reports as well as the evaluator’s interviews with physicians. NORC-SSP staff, across all sites, periodically shared information with their Linkage health partners via telephone, fax, mail and in-person visits. Over time, communication became a two-way street (see Table 4), providing evidence that communication was taken to a new level. All sites reported that communication that transpired across settings became more proactive in nature over time. Health providers began reaching out for assistance or information as opposed to only responding to NORC-SSP-initiated requests. One goal of increasing awareness and communication was to increase identification of residents in need. Theoretically, increased identification of need would lead to improvements in continuity of care. Table 5 presents

evidence demonstrating how the Linkages resulted in greater attention to residents’ needs in the form of enhanced tracking across settings, improved recruitment, and increased service provision. Across the sites, preliminary evidence of increased shared planning was found (Table 6). At some sites this translated into increased referrals among providers. At other sites, this translated into use of shared information for diagnosis or treatment planning. For example, at the Lower Manhattan site, staff reported that prior to the Linkage project there was little interaction between the NORC-SSP and the primary care center. Indeed, the NORC-SSP was unsuccessful at getting the primary care center to respond to inquiries or referrals. However, once the NORC-SSP focused its efforts on diabetes disease management approach, the primary care center took notice, recognizing that this kind of cross-sector collaboration would enable them to extend their effort into the community, a key component of chronic disease management. Soon, the primary care clinic began recruiting NORC-SSP residents to become a part of its ongoing Diabetes Collaborative which included group visits, treatment plans, shared care planning and information exchange. At the Queens site, residents reported that their doctors were not routinely asking them about falls, nor were clients themselves sharing such information. Once assessment findings and care plan recommendations for residents found to be at-risk were sent to the residents’ physicians, an increasing percentage of residents reported having more discussion with their physicians about mobility issues. Participating residents also began to present their physicians with a chart sticker indicating their participation in the collaborative falls program which helped to initiate a conversation about falls or falls concerns. Interestingly, although the MyMeds profile was reported to add value to the care planning process for residents,


10


Table 4. Increased communication among partners Site

Evidence

MidManhattan

• On average 117 contacts per quarter, 28% related to medications • On average 8% of all contacts with the NORC program were initiated by a pharmacy per quarter

Lower Manhattan

• Eighty-four patient-related contacts documented for the 39 test patients; of those, 15 were referrals from the primary care center to the NORC program, 12 were new referrals for home care services, and 15 involved self-management plan interventions

Queens

• Approximately 30 contacts (telephone, fax, in person) between the NORC-SSP and community physicians per quarter; increasing percentage initiated by the community physicians (20% on average, per quarter), • NORC-SSP communicated with 39 different community physicians during the project

Brooklyn

• Over 50 patient-related contacts per quarter (over 200 by the end of the implementation year for 45 enrolled patients) with 34 participating community physicians • Increasing number initiated by community physicians (11% by the end of the implementation year)

Table 5. Increased identification of residents in need Site

Evidence

MidManhattan

• One hundred percent of ED visits by participating residents enrolled in the MyMeds program were reported to the NORCSSP • Approximately 30% of the (100) enrollees were new to the NORC-SSP

Lower Manhattan

• All enrollees received at least one intervention (e.g. home care referrals, NORC nurse visit, phone reminders, etc.) through the integrated diabetes assessment program, with many receiving multiple interventions

Queens

• Approximately 100 client assessments or reassessments were conducted by NORC-SSP staff (approx. 25 per quarter) using the Hartford Falls Risk Assessment protocol • Approximately 70% of those clients assessed for falls risk were determined to be at risk and in need of intervention

Brooklyn

• Approximately one-quarter of all participating residents reported that their physician or physician office staff asked if they were a client of the Brooklyn-based NORC at the time of an office visit

Table 6. Increased shared care planning Site

Evidence

MidManhattan

• Providers reached out to the NORC-SSP to get information on shared patients or to find out how to enroll other patients into the Linkage project

Lower Manhattan

• Fifty-two collaborative assessments of diabetes status conducted for the target patients over the course of the year • Quarterly group visits at the primary care center drew approximately 5 NORC program clients each time • Fifty-six percent of residents report seeing the NORC-SSP and the primary care center staff work together to assist them in their diabetes treatment

Queens

• Providers responded to assessment findings and care plan recommendations that were sent by the NORC-SSP nurse or presented by the resident • Residents reported that physicians began asking them about mobility issues and falls

Brooklyn

• Twenty-four percent residents reported that their physicians either looked or wrote in the client passport

pharmacies and the Mid-Manhattan NORC-SSP— indeed community physicians who learned about it through their patients began requesting information for their other patients—it went unutilized at the ED. Further investigation of why the ED providers did not utilize the profile should be conducted, especially since they were quite enthusiastic about the profile during the planning phase.

Additional anecdotal evidence gathered at all sites suggests increased continuity of care (Table 7). At the Mid-Manhattan site, NORC-SSP staff reported that the sharing of admission and discharge information resulted in greater attention to medication and other health-related issues by program staff and residents. Towards the end of the implementation year the Mid-Manhattan site expanded their


11


Table 7. Increased continuity of care Site

Evidence

MidManhattan

• Shared medication information led to sharing of other health information across providers • By the end of the implementation year, 14 pharmacies had joined the MyMeds network

Lower Manhattan

• All diabetes linkage program enrollees visited a doctor during the implementation year • Forty-eight percent missed appointments in 2005, only 27% missed appointments in 2006, a 45% reduction

Queens

• By the end of the implementation year, the majority of residents surveyed were able to identify falls risks • Increased number of residents report telling their physician about a fall (38% in the 1st quarter, 50% in the 4th) • Increased number of residents made a change to prevent a fall (53% in the 1st quarter; 67% in the 4th) • Fewer participants reported a fear of falling (35% in 1st quarter, 20% in 4th) • Fewer participants reported discomfort speaking with providers about falls

Brooklyn

• Increased comfort talking to physicians about emotional health issues (from 12% in 1st quarter to 43% in 4th) • Increased comfort talking to NORC-SSP staff about emotional health issues (from 27% in 1st quarter; 47% in 4th) • Increased numbers of clients reported showing their passports or NORC-SSP chart stickers to physicians over time (10% in 1st quarter; 30% in 4th)

MyMeds intervention to include medication education and adherence monitoring for those determined to be at risk. Approximately 60% of new enrollees in the last two quarters of the implementation year received some sort of medication education, including: dosage clarifications, education about what the medications were for, medication reminders, monitoring; and, attention to changes in medications and health status. The MyMeds resident bracelets that were created to better identify Linkage participants turned out to be extremely popular among residents and providers. Residents began requesting additional information (e.g. diabetes status, allergies) for the bracelets. Staff report that having such information on hand made residents feel more confident that there would be ready access to current information if they should need it. Furthermore, residents reported that the bracelet helped them better communicate with physicians. Providers viewed the bracelet as a way to minimize errors and increase continuity of care. Furthermore, the bracelets helped the NORC-SSP communicate with providers, recruit new participants (other residents began inquiring about the MyMeds bracelets) and health partners (14 new pharmacies joined the network by the end of the implementation year), and identify enrollees at the point of hospital admission, pharmacy contact or community-based physician office visit. At Lower Manhattan, the Linkage led to improved primary and preventive care as well as visit adherence— both indicators of improvements in continuity of care. Furthermore, evidence suggests that clients were taking a more proactive role in their care, which is a precursor to improved care continuity. By the end of the implementation year, 100% of enrollees reported

awareness of risk factors for diabetes, 77% reported awareness of proper foot and eye care; 100% reported enhanced confidence in managing their diabetes (up from 65% in quarter 1); and, 100% enrollees had a primary care center-developed Self Management Plan (data not shown). Access to health information enabled the NORC-SSP to track health trends within its target population for the purposes of health promotion planning. Indeed the diabetes linkage initiative has led to additional joint programming focusing on other health conditions at this site. At the Queens site, improvements in resident knowledge of falls risk and self care provides preliminary evidence of improved care continuity. Approximately 20 lectures or health promotion activities were provided to participating residents over the course of the Linkage implementation year with an average attendance of 20 residents at each. Falls-related information was included in the quarterly newsletters; over 1400 newsletters were distributed each quarter. Furthermore, staff report that the adoption of care plan recommendations made by the Queens NORC-SSP to community physicians about residents’ falls risk provides evidence of shared care planning as well as increased continuity of care. Indeed, the Queens site witnessed physician follow-up on NORC-SSP suggested recommendations and subsequent physiciandriven referrals to implement recommendations (e.g. home care, physical therapy, and counseling, ordering of durable medical equipment). At the Brooklyn site, evidence of increased resident self-advocacy represents a critical step towards improving continuity of care. Despite the short time period, evidence of the more distal care outcome was found across sites (Table 8). At the Mid-Manhattan site, preliminary evidence


12


Table 8. Improved care outcomes Site

Evidence

MidManhattan

• Number of ED revisits at 30 days decreased from the beginning of the implementation year to the end, from 21% of all ED visits to 5%

Lower Manhattan

• By the end of the implementation year, enrollees were showing improved levels of A1c (14% increase in number of enrollees with A1c